Dr. Sharp from the Koch Institute for Integrative Cancer Research @ MIT. Visiting scientist – Nobel Prize winner, among many other awards.
Small RNA and cancer biology. How small RNA fits into gene expression of small cells. Cancer IS a disease of gene regulation. You can look at cancer and try to understand it in terms of regulation.
With the discovery of RNAi, we began thinking of RNA as a regulatory factor. Something about Pi RNA? (I have no idea what Pi RNA is. – must look this up.) Now we also have to contend with lincRNA. Fortunately High Throughput sequencing gives us the tools to investigate this.
MicroRNA pathway overview. PolII -> miRNA gene -> Drosha produces hairpin -> Export to cytoplasm -> Dicer cuts it -> Argonaut binds and carries to mRNA -> regulation or repression.
Regulation happens for partial complimentation, repression happens when complimentation is perfect.
Dicer is the only enzyme that does the cleavage of hairpins. Only argonaut2 gives you cleavage of mRNAs.
8-nt Seed region is the most important part for binding.
- Most target sites are in seed sites in 3’UTR.
- Evidence for miRNA targeting through “centred paired” sites 4-14 or 5-15
- Most protein silencing is less than 1.5 fold, but can be synergistic up to 10-fold
- Silencing can also happen – most cases mRNA degradation, but sometimes translation
- Little evidence of miRNA activity in nucleus. (up for debate)
- 87 evolutionary conserved “seed families” of miRNA
- 1/2 of all genes have no 3′ UTR.
- 1/2 of all genes have one or more target, average of4.2, 200-500 target mRNAs per family.
UTR complexity: – high, see examples of HMGA2, BIM, CASP2.
Transcription regulatory control – Reprogramming of somatic cells to pluripotency. (Oct4, Sox2, Klf4, cMyc). TFs are master regulators – control cell state, all other processes are subservient to them. Discussion of “circuitry of embryonic stem cells”. The master TF’s for this process also drive miRNAs.
For most cell types, a few types of miRNA dominate. ES cells have mir-290-295, Embryonic fibroblasts: mir-21 and let-7. Neural Precursors, let-7.
miRNA circuitry in S cells – incoherent feed forward regulation. (Nice images of pathways…) You get strong induction of protein levels, but also tight regulation. (Strong promoting while also creating strong repression with miRNA.)
So: Can a decrease in regulation by miRNA promote cancer? Contend yes.
See Hanahan & Weinberg, 2000. Great picture of oncogenes and where miRNAs fit in. eg, Let-7 controls Ras, mir17-92 paired with myc, mir21, mir15 and mir16 control/repress Bcl2. Etc.
- mir-15a and miR16-1 -> CLL prostate
- Mir26-a -> Liver
- etc etc…. too many others.
How does this work? Kumar and Jacks tested this with a mouse model system: activating Kras, repress p53 (flox), and Dicer(flox) repression. Does loss of miRNA inhibit activity of tumour?
- Dicer (+/+) = 112 days survival
- Dicer (+/-) = 87 days
- Dicer (-/-) = 101 days. (something about heterozygous dicer presence.??)
[This sounds like a very neat experiment, but I’m not following, since I don’t know what cree/flox are. Must read up on it.]
Dicer null cells retain the ability to form tumours. miRNA are not essential for growth OR for cancer.
Roll of miRNA regulation:
- Stabilise quiescent state
- Stabilise differentiated state
- Control responses to changes in environment-induction of cell death.
Cells lacking miRNAs are more susceptible to stress. (FACS data.) Increase in Caspase3 when miRNA is knocked out, cell death.
Thus, you get more stable on-off transitions, and better stabilisation of expression.
Examples given with a reporting cell line. Straight line expression becomes hockey stick like when miRNA sites are added. Thus, suppressing expression at low levels, but not at higher levels.
All of this provides a more sensitive transition – better on-off transitions. Helps act as a buffer against leaky expression. You get a more sigmoidal gene expression curve, which significantly helps the cell control it’s states.